🖤 Greyscale CNN Project – Fashion MNIST Classification

A hands-on project that demonstrates building and training a Convolutional Neural Network (CNN) on grayscale images using the Fashion‑MNIST dataset (built into TensorFlow/Keras).

---

🚀 Project Overview

This project showcases a complete deep learning workflow:

• Loading and preprocessing Fashion‑MNIST grayscale images (28×28 pixel format)
• Constructing a CNN (Conv → Pool → Dense architecture) using TensorFlow / Keras
• Training and evaluation with accuracy and loss curves
• Visualizing sample predictions and misclassified examples

Built as a Jupyter Notebook (Fashion_Mist.ipynb) for interactive learning and reproducibility.

---

📂 Repository Structure

├── Fashion_Mist.ipynb # Jupyter Notebook for the full ML pipeline ├── .github/workflows/ # CI or automated scripts (optional) ├── .gitignore # Files to exclude from Git └── README.md # Project overview guide

---

🎯 Key Features

• ✅ Load and preprocess grayscale image dataset via TensorFlow datasets
• ✅ Build CNN architecture with convolution and pooling layers
• ✅ Train and evaluate models using accuracy and loss metrics
• ✅ Visualize results: showing actual vs. predicted labels
• ✅ Modular, educational notebook format ideal for duplication

---